LTSPICE bug?

[Mosaic]

Hi All:
I can't seem to get LTspice to simulate this simple circuit:
https://hackaday.com/2011/12/06/reverse-voltage-protection-with-a-p-fet/

It won't shutoff the PFET with the supply on the drain and the load on the source.
Even if I tie the gate to VCC.

Any thoughts?

 

[alec_t]

The body diode shown passes current regardless of the gate voltage. Reverse the FET if you need to shut it off (but then the circuit won't function as per the linked article).

 

[Daniel Wood]

Looks fine to me..
Its probably best to use a "real" device instead of the default unassigned PFET. I've had issues simulating with generic transistors and fets.

 

[crutschow]

It won't shutoff the PFET with the supply on the drain and the load on the source.
Even if I tie the gate to VCC.

The simulation is fine.
The bug is in your test.
The circuit is not designed to turn off under gate control.
It's designed to block a reverse voltage to the load, as Daniel's simulation shows.

 

[Mosaic]

Ok, I see, in order to achieve gate control I'd need back to back PFets, one for polarity blocking and one for gate. Cuz of the body diode, not shown in the LTspice schematic.

 

[MikeMl]

... the body diode, not shown in the LTspice schematic.

 

[Mosaic]

That indicates if it is a PFET or an NFET... that's not the body diode dude. The body diode sits across the drain/source.

But then maybe I'm wrong about that.

 

[MikeMl]

The multitasking arrow also indicates the direction that the device conducts when the Voltage across the Drain-Source is reversed biased (compared to normal NMOS or PMOS polarities), so the arrow implies current flow from Drain to Source when the drain is positive and source is negative in a PMOS.

 

[throbscottle]

That's a really good tip about the arrow, Mike! I'd finally settled on "substrate connected to source" as my mnemonic for remembering which way up these symbols go ('cuz they both begins with "S"), but looking at it as the body diode is much better! Thanks